The application's broad, long-term objectives are to discover a blood test for early detection of ovarian cancer that will reduce ovarian cancer mortality through regular testing of targeted populations. Initially these populations would include women at high risk due to family history and/or presence of a BRCA1or BRCA2 mutation within the family, and all postmenopausal women where the incidence of disease is highest. The test requires high sensitivity for early stage disease and very high specificity so that few false positive tests will occur for each true positive test. The specific aims are 1) to discover high probability candidate biomarkers through proteomic analysis of biofluids from three sources a) ovarian cyst fluid from benign and malignant ovarian disease, b) conditioned media from fresh sliced washed normal and malignant tissue, and c) conditioned media from phenocopy model fallopian tube systems, 2) discover high probability candidate biomarkers from genomic analysis using Affymetrix arrays of normal fallopian tube, normal ovarian, and ovarian malignant tissue lysate filtered by a comprehensive list of secreted proteins from ovarian tissue (secretome), 3) prioritize candidate biomarkers for verification by analysis of pathways from ovarian cancer pathogenesis, 4) construct mass spectrometric assays for the top 50 candidates and measure these candidates in plasma from 100 cases and 100 benign controls, and in longitudinal plasma in cases prior to clinical detection, and in controls from an ovarian cancer screening trial, and 5) determine which candidates have earliest sensitivity by estimating the change-point (if any) at which the candidate rises significantly above baseline. The five best candidates will form a biomarker panel for further testing and refinement, outside the scope of this application, in the biorepositories of larger scale screening studies.